enhanced NB-IoT: Disrupting the cellular IoT supply chain

July 9, 2018 Brandon Lewis

By now you’re certainly aware that the vast majority of Internet of Things (IoT) devices will be low-power sensor nodes distributed at the (sometimes far) edge of networks. These systems are being deployed in applications from wearables and remote surveillance to environmental monitoring and asset tracking, and emphasize low cost, long battery life, indoor coverage, and comparatively low bandwidth. These requirements are often at odds with traditional wireless networking options.

In response, industry has invested heavily in low-power, wide-area (LPWA) network infrastructure in recent years. The growth of these technologies has been significant, as shown in Figure 1, with cellular IoT solutions projected to grow at a 30 percent compound annual growth rate (CAGR) from 2017 through 2023.

Figure 1. Global eNB-IoT and LTE Category M1 connections are forecast to grow by 30 percent year-over-year through 2023.

Most of the growth referenced in Figure 1 will be from enhanced NarrowBand IoT (eNB-IoT), Release 14 of the 3GPP’s NB-IoT standard. eNB-IoT repairs a flaw in Release 13 of the standard that resulted in NB-IoT chips drawing too much power, and also improves sleep/wake state switching to minimize power consumption. Most importantly, eNB-IoT more than doubles the data rate of Release 13 from 60 Kbps to 160 Kbps, which serves to further reduce power and prolong battery life (up to 10 years) because power amplifiers (PAs) spend less time working to transmit data.

Every NB-IoT operator and chipset manufacturer will be commercially deploying Release 14 eNB-IoT solutions moving forward.

As opposed to non-LTE narrowband LPWA networking options like LoRa and Sigfox, eNB-IoT offers improved quality of service thanks to operation in licensed spectrum bands, deterministic latency, higher data rates, and true bi-directional communications. Compared to other LTE IoT connectivity standards, such as Category M, eNB-IoT provides a much simpler PHY, communications stack, and certification process at half the cost and one-third the power consumption.

Disrupting the IoT network value chain

With eNB-IoT technology lowering barriers to entry, minimizing risk, and decreasing time to market, the IoT network supply chain is experiencing a period of disruption. For example, chipmaker Nordic Semiconductor recently announced the nRF91, a dual-mode Cat-M and NB-IoT system-in-package (SiP) that functions as a cellular module. Meanwhile, module maker u-blox just released the NINA-B2 series that contains a chip developed in house. New competitors with no prior cellular experience are also entering the cellular IoT market in the hopes of delivering differentiated silicon solutions. All of these players are shooting for a target price point of $2 with chipsets that include a modem, PA, applications processor, sensors, etc.

To foster this disruption, CEVA released the second generation of its DragonFly wireless communication IP – the DragonFly NB2 – which is the first eNB-IoT Release 14-compliant IP stack (Figure 2). In addition to eNB-IoT, DragonFly NB2 incorporates multi-constellation GNSS support and a voice processing software front end. The complete stack includes:

  • Complete eNB-IoT GNSS digital front end (DFE), RF transceiver, and 14 dBm PA
  • Modem (integrated in software)
  • Power management unit (PMU)
  • CEVA-X1 multi-purpose DSP and control processor
  • USIM and eSIM interfaces
Figure 2. CEVA’s DragonFly NB2 is a complete eNB-IoT and GNSS IP stack.

On the software side, the DragonFly NB2 integrates the eNB-IoT Release 14 UE protocol stack, Layer 1 PHY, encryption libraries, platform drivers, and the freeRTOS operating system. CEVA also developed new instructions for the DragonFly NB2 allow the clock speed of the CEVA-X1 processor to be lowered, and help reduce MIPS and power consumption for intensive modem and GNSS tasks by a factor of eight versus the DragonFly NB1. In fact, these instructions have been demonstrated to be more power and area efficient than dedicated hardware accelerators.

Other intelligent mechanisms, such as time-division multiplexing of software tasks, ensure a consistent ultra-low sleep power consumption of a few mA (Figure 3).

Figure 3. Time-division multiplexing on CEVA DragonFly NB2 IP allows the processor to stay in ultra-low sleep mode during which it consumes only a few mA of power.

DragonFly NB2 IP is silicon-proven, and has been manufactured at 40 nm and 55 nm process nodes in 10 mm x 10 mm chips. DragonFly NB2 can be licensed as a whole IP stack or just a subset.

eNB-IoT IP: Flexibility for the future

The IoT wireless landscape remains in a state of flux as new technologies are introduced and others that were rushed to market (like 3GPP’s NB-IoT Release 13) are refined. For those looking to enter a market primed for disruption without incurring the risk associated with volatile trends in wireless networking, IP solutions such as the DragonFly NB2 offer enough flexibility through software to protect investments. For instance, CEVA’s DragonFly NB1 solution was developed for NB-IoT Release 13, but can still be upgraded with an eNB-IoT Release 14 protocol stack.

In all, comprehensive low-power IP stacks can help the changing IoT wireless network supply chain get to market quickly with differentiated products that are more cost-effective than off-the-shelf silicon alternatives.

Reference chips and a DragonFly NB2 Silicon Development Kit will be available for evaluation in Q3 (Figure 4).

Figure 4. Available in Q3 2018, the DragonFly NB2 Silicon Development Kit includes the CEVA-X1 processor, baseband, and RF on a single die, supported by UART, SPI, GPIO, and other peripherals. Debug is supported over JTAG or UART (RTSS), with software development tools and documentation provided as well.

About the Author

Brandon Lewis

Brandon is responsible for Embedded Computing Design’s IoT Design, Automotive Embedded Systems, Security by Design, and Industrial Embedded Systems brands, where he drives content strategy, positioning, and community engagement. He is also Embedded Computing Design’s IoT Insider columnist, and enjoys covering topics that range from development kits and tools to cyber security and technology business models. Brandon received a BA in English Literature from Arizona State University, where he graduated cum laude. He can be reached by email at blewis@opensystemsmedia.com.

Follow on Twitter More Content by Brandon Lewis
Previous Article
Rapid response for real-time control
Rapid response for real-time control

Leverage core-independent peripherals to create more efficient embedded systems

Next Article
Cryptography-enabled microcontroller with DICE architecture simplifies development of secure connected nodes
Cryptography-enabled microcontroller with DICE architecture simplifies development of secure connected nodes

New development kit for Microsoft Azure IoT Hub makes creating secure connected devices easy

×

Stay updated on processing and related topics with the Processing edition of our Embedded Daily newsletter

Subscribed! Look for 1st copy soon.
Error - something went wrong!